F4 with -4 Empenage

[Xkuzme1]

Hey y’all. Question. What exactly is the limiting factor for Vne on the Rockets. Tail feathers or wings?

If an F-4 has an rv-4 set of tail feathers does it still have the 212 tas Vne? Or would it be higher?

What about the balanced rudder. I’ve actually heard a pretty good argument that the balanced rudder is more likely to have flutter than the -4 trailing rudder.

Thoughts?

X

 

[sailvi767]

I once owned a very fast RV6 with a balanced rudder. It was based off Dave Anders design. It was a small but balanced rudder and tested to 230knots. It was built by Robbie Attaway. He was not comfortable with the stock rudder at speeds above the published numbers. I believe the original Harmon used the stock RV4 tail with some reinforcement. Others can comment more.
As far as the wing does the aircraft have the standard RV4 wing or the shortened Rocket Version?

 

[RV8JD]

What about the balanced rudder. I’ve actually heard a pretty good argument that the balanced rudder is more likely to have flutter than the -4 trailing rudder.

I'd like to hear that argument.

 

[vfrazier]

What exactly is the limiting factor for Vne on the Rockets. Tail feathers or wings?

Great question. Sadly there is no single answer since these aircraft, whether RVs, F4s, or Rockets, are all built by people with varying skills and incorporating infinite customization.

If you're building an F4, I recommend going with the least common denominator, so to speak. If Van's says the VNe is 212 for the RV-4, then if you're using RV-4 parts, anything past that makes you a test pilot.

For the record, I'm not an engineer, but I've been blessed to work with many, many of them since 1983. It's humbling, and eye-opening. If you can get any of the several engineers on this list to bless your plans, that is a great start, but you still need to flight test to prove it.

For this discussion, as it is basically a question of busting VNe, I am aware of many, many incidents over the years where pilots have asked their aircraft to do more than it was capable of. Not all of these stories have a happy ending.

I would suggest to anyone who wants to make their aircraft go faster to do this: For every hour you spend thinking of (and engineering) ways to make your aircraft go faster, spend 2 hours pondering the definitions of maneuvering speed, VNe, and how close you are getting to the edges of the envelope.

Just because John Doe made his Whizbang 2000 set a record at Reno doesn't mean that your aircraft, with you at the controls, is capable of doing the same thing safely.

Truth. Don't shoot the messenger.

 

[RV8JD]

I would suggest to anyone who wants to make their aircraft go faster to do this: For every hour you spend thinking of (and engineering) ways to make your aircraft go faster, spend 2 hours pondering the definitions of maneuvering speed, VNe, and how close you are getting to the edges of the envelope.

And I'll beat Scott M. ("rvbuilder2002") to it, but add Vno (Max Structural Cruising Speed) to the list.

 

[vfrazier]

+1 for what Carl said.

 

[I-TERA]

I fly an F1 rocket evo.
I had rudder flutter, which by pure chance ended up being good for me and the aircraft. This prompted me to study the problem. First read AC23-629-1b. The simplest approach, within the reach of anyone who wants to do a flutter risk assessment, is to apply the old Report 45, not sufficient to ensure that the experimental result matches the calculation, but very useful. It is limited to speed up to 300Mph.
By studying it, one understands that there is no rudder that is better than another or that balancing is useful or not, one understands that the dynamic equilibrium, not the static one, of a mobile surface must be associated with the mode and frequency of oscillation of the rest of the plane. For example, the natural frequency of fuselage torsion directly determines the speed at which a given rudder will oscillate divergently. Increasing the rudder counter mass could be negative or could be useful, it doesn't depend on the rudder itself but on the behaviour of that aircraft with that rudder.
I think it is very dangerous to experiment with high speeds without calculations and at least a ground vibration test.
The behavior of the association between a fuselage of an aircraft that has a Vne of 250Kn and a rudder of another aircraft that has a Vne of 250Kn is not easy predictable, it could be unstable at 200Kn!

 

[sailvi767]

Do you mind giving us a brief on your rudder flutter issue and the speeds involved?

 

[I-TERA]

During the flight envelope tests the speed reached in still air and straight flight was 270Kn with no signs of instability.
A turn with just over 1.5g at 238Kn showed instability, with torsional deformation of the rudder.
After the calculations and the modifications made, the ground vibration tests confirmed the absence of flutter up to 311 Kn. ( 160m/s at 8000' )
The flight tests up to 160Kn with strong stresses on the controls and up to 270Kn with small stresses gave positive results.( up to 8000'. The aerodynamic damping and consequently the probability of flutter depends on air density, so variation in temperature or altitude can surprise you...)
A triaxial accelerometer applied to the empennage during flight tests measured the damping of the oscillations caused by the pulses on the controls.
I consider the entire flight test procedure performed a little risky due to my lack of pilot experience and skill and the simplicity of the equipment used.

 

[PaulRomano]

Claudio, I read your engineering report several years ago, and I thank you for it. I fully agree that a GVT along with the correct analysis software can help in understanding a problem of this kind. But as you point out, very careful flight testing is also necessary to validate the results.

We have had small number of reported cracks in the aft fuselage. We are assuming (but this could be an incorrect assumption) these aircraft have developed these cracks while maneuvering at high speeds. At 240KTAS a Rocket stabilizer can develop approximately 1000 lbs of load (per side) at 5-7 deg AOA. The internal reaction loads can be much higher. Critical portions of the internal structure will be close to yielding, and while the structure may not fail, the probability of cracks forming will be higher.

Over the years various well intentioned fixes have been added that have increased the weight of the aft fuselage, and in addition the F1 empennage. These fixes ripple through other aspects of the design and at a minimum reduce cg range.

Your GVT analysis may have shown an absence of flutter to 311KTAS, but that does not mean the underlying structure can sustain maneuvering loads at that speed. I believe you and your team in Milan understand this, but others may not.

Please be careful, and I would have enjoyed working with you and your team while you were doing this study. And thanks again for sharing your results.

 

[vfrazier]

Your GVT analysis may have shown an absence of flutter [at speeds approaching or beyond VNe], but that does not mean the underlying structure can sustain maneuvering loads at that speed.

Words in bold are mine for emphasis as this topic applies to all of our aircraft. This post isn't intended to apply to any specific RV or Rocket type aircraft as they are all reasonably similar.

This is something that should be considered. Flutter may occur, and we should do due diligence to avoid, or correct, it, but flutter isn't required for a part to fail. Simply overloading, or fatiguing a part can fail it.

As Paul pointed out, the HS on our planes can see a substantial load at higher speeds. Recall that the HS needs to produce a down force to trim the lift that the wing produces. In general, more speed means more down force.

I'd challenge anyone to test their HS by having 4 full size adult males stand on each side of their HS. The HS, in theory, won't likely fail, but what if you repeat that load dozens, hundreds, or thousands of times. That is what may be happening to some of our faster-than-intended aircraft when they fly in even modest turbulence, or are maneuvering.

We all want to go fast(er), but give that need for speed the respect it demands. These aircraft have a margin of safety built in. It's your insurance policy. At some point, that margin becomes zero. Don't intentionally go there.

YMMV

 

[sailvi767]

I am curious about this. Should not the downforce load on the tail be the same regardless of airspeed at the same G? If you pull 6 G’s at 160 knots is that less load than 6 G’s at 200 knots?

 

[vfrazier]

George,

Full disclosure, I am not an engineer. If it is pointed out that my generalized comments are incorrect, I'll gladly take them back. My whole intent in this thread is simply to caution against thinking that our aircraft can simply be made to go faster, and faster without careful consideration of the margins of safety that were engineered into them.

As Paul Romano mentioned in the post above yours, we are gathering info related to the operation of our aircraft with the end goal of learning how/what/where/when/why are we occasionally seeing cracks, or other problems. We welcome discussion and feedback and invite F1, F4, and other Rocketeers to visit our display at Oshkosh this year to chat with us. Space 645, just East of the forum buildings, and just SW of the Van's display.

So, your question was:

I am curious about this. Should not the downforce load on the tail be the same regardless of airspeed at the same G? If you pull 6 G’s at 160 knots is that less load than 6 G’s at 200 knots?

The pilot will feel the same load, but various parts of your aircraft may experience differing amounts of those Gs. This thread began with a question about VNe and tailfeathers. Since there have been few, if any, wings or engines falling off, we'll concentrate on the tails. There have been incidents related to the tails, but thankfully, no accidents... that I'm aware of (for the F1s or F4s).

Large transport, and other, aircraft typically have a moveable HS. The HS is adjusted to trim the aircraft at various speeds, loading, etc. (or they can move fuel to adjust CG, or other magic tricks, to maximize efficiency). In theory, the loads on their properly trimmed HS should remain relatively low, i.e. only as much load as is required to maintain whatever flight condition they are in. Generally speaking.

We are limited in what we can do here, as the incidence of our HS is fixed. Aerodynamic loads react on it depending on airspeed, although AoA of the main wing, CG, aircraft weight, etc. all have an effect. When we trim our aircraft, it is by moving the elevator, not by changing the HS incidence. Therefore, in general, the HS sees the aerodynamic lowest loads at low speed/high AoA, i.e. landing, and sees the highest loads at high airspeed, although elevator position certainly does transfer load into the HS rear spar.


Generally speaking:

• no one is breaking anything at low airspeed, but we have seen cracks, and a few incidents at higher speeds.*

• Flutter, while a bad thing, isn't necessary to break something.

• Since there is already down force on the HS, at higher speeds pulling on the stick just adds to that downforce. This is why we're discussing that safety margin! At some point, there is no margin left to pull on the stick.

As I said earlier, We all want to go fast(er), but give that need for speed the respect it demands.

YMMV

*Consider the various service bulletins related to cracking that have been issued by Vans, and others.

 

[PaulRomano]

George, great question, and I will answer it in 2 parts.

But the quick answer to your specific question is that the force generated by a 6 g pull at 200 or a 6 G pull at 160 is the same. It is 6 multiplied by the weight of the aircraft.

First, we are talking about non accelerated flight, which translates to no turning, and 1 g. Straight and level, descending, or climbing at a constant speed. The stabilizer is a small wing, that needs to operate well at both low and high speed. The lift it generates is a function of the speed squared, the area of the stab, and the elevator deflection (or the lift coefficient). At low speed, we use more elevator deflection to land because we need to trim the aircraft during the flare. At high speed, very little elevator deflection is needed to generate the same lift force.

At Rocket Vne, simply plugging the numbers into the lift equation shows that at modest elevator deflection the loads can get very high. And this is at 1g (non accelerated flight).

G's are incurred when the aircraft changes direction. Pull back on the elevator, then neutralize the stick, and the aircraft sees a certain g force when the elevator deflects which depends on the magnitude of the elevator deflection (which affects the turn radius) and the the aircraft speed squared. Assuming you do a quick pull to 10 deg first at 160, and then again at 200, you will generate 60% more G. These Gs, multiplied by the aircraft weight, become forces on the wing and tail that are added to the forces described above.

A classic unintentional scenario: IFR flight plan, but VMC, congested airspace, center requests an expedited descent from 10 to 8 with a heading change. Throttle back, start descent and turn and the you suddenly realize you will bust 8. You are Vne and already at a problematic airspeed. Gently pull back the stick to arrest descent and momentarily pull 2 Gs. And suddenly there's a thump and a sound heard.

And all this happened in still air, no turbulence or gust loads.

When aircraft are designed for a specific G load, it is also at a specified weight and speed (Va). And it's important to keep all 3 as well as Vno in mind.

 

[sailvi767]

I have always had some concern about the Rocket VNE. I set the VNE on the Dynon for 225 KTAS. I have had zero issues with that speed and just don’t see the need for 239 KTAS. I did solo testing to slightly past VNE of 239 after the purchase with no issues but you have to work hard to get the aircraft to that speed. 225 provides everything we need to fly the aircraft even in descents from the teens. The airframe is pretty draggy at high speeds. A very modest power reduction will give you a 1000 FPM decent.

 

[I-TERA]

Hi Vince, Paul.
You both are right.
I also think that Vince's new empennage is much better than the original one.
The VNE of the Rocket Evo is suggested at 250Kn, the test up to 1.1 Vne in still air, with a speed increase of less than 1 knot per second, is what seemed to me closest to the Far 23 that my courage allowed.
So after the tests I decided to reduce the Vne of I-TERA to 240Kn, accordingly to the speed reached.
I have no data for the Vdive, nor for the maximum structural speed.
A rough calculation did not show dangerous weaknesses for the empennage, it should be considered that I strengthened, to the detriment of the Cg ( but the battery is further forward than the spar), both the stabilizer and the attachment to the tail cone.
The load on the stabilizer and therefore on the empennage structure is already high at 200Kn, without trim the joystick is very heavy, for this plane particular attention must be paid to the trim run-away, at low altitude ( low pass) you may not have the possibility to recover.
My airplane cruises at 8000-9000' between 210 and 220Kn, the speed is very bank account dependent...here in EU 1 liter avgas is 2,9 € !!!
At cruising speed, maneuvers must necessarily be smooth.
I think that Mark's suggestion, the stick is not a wing remover, is wise.
The reason for my intervention was to suggest extreme prudence in making changes, even small ones, without the appropriate evaluations.
It should also be remembered that, as far as flutter is concerned, the Vne is Tas and not Ias.
Be careful.
Claudio

 

[vfrazier]

George, Claudio, and others,

I think we're all trying to express much the same concerns here. Thank you. These are the things that would keep any sane person awake at night!

The reason for my intervention was to suggest extreme prudence in making changes, even small ones, without the appropriate evaluations.

 

[Xkuzme1]

Do you mind giving us a brief on your rudder flutter issue and the speeds involved?

Just purely based on a “from the hip” observation that most planes that have experience flutter were balanced rudders. It’s not scientific. Just an observation. It came from a friend that has built several RVs, and is a very experienced pilot and mechanic.

X

 

[Xkuzme1]

And don’t get me wrong fellas. I’m not looking to push any limits. But in order to not push them, I want to understand them and know what exactly they are.

I understand that if I have a concord with RV-4 wings, then the wings are the limiting factor. No Mach speeds today.

I am simply trying to complete a build that someone else started. It’s a little bit of a detective job at this point. Some of the items were completed in a non-standard fashion. It’s easier to decide to change parts now than it would be later if unrealistic limitations were realized.

X

 

[vfrazier]

But in order to not push them, I want to understand them and know what exactly they are.

Bold added for emphasis. Personally, I don't believe that it is possible to know this information. Even the best engineers on the planet struggle mightily towards this goal.

I mean, if Boeing can build a 737 MAX, then what possible way is there for any of us lesser beings to expect perfection? Boeing arguably stretched an older design past it's original limits and got bit. Should we expect anything different?

Let's sum this discussion up graphically.

No homebuilt kit, parts, or aircraft manufacturer can possibly know what any builder or pilot will do with their products. The whole point of my responses is simply to ring the bell to make folks aware that they must have realistic performance goals.

If you have the "need for speed" above all else, then you must assume the risk.

 

[I-TERA]

https://www.dropbox.com/scl/fi/no2f...t-45.ods?rlkey=g3faajh4eijgwdstx02wu747w&dl=0

This is the link to the spreadsheet I used to evaluate which counterweight was right for my rudder.

Unfortunately there is no short way, at the very least, and the Report 45 is the minimum, you have to measure the natural frequency of bending and torsion oscillation of your fuselage. The stiffer the fuselage, the higher the resonance frequency, the faster you can fly without empennage flutter.
The spreadsheets bending and torsion frequency are those of my F1 Rocket.
The rudder is also the rudder I made and you will find the relative moments of inertia in the spreadsheet.
With my fuselage and my rudder you can try to change the counterweight, until it is reduced to a minimum.
You will see that the maximum acceptable torsion and bening frequencies also change.
Try changing the maximum speed.
You will see that the frequencies also change in this case.

If the rudder is symmetrical with respect to the torsion axis, see Macchi Castoldi M.C. 72, 1934, a rudder without a counterweight can fly at 730Kmh. Or you have a rigid fuselage like the P-51.

The Report 45 is old, it doesn't solve every aspect of the flutter problem and therefore it doesn't exactly guarantee the result, but it doesn't take into account the aerodynamic damping and this makes it a little too cautious, which is fine for us amateur builders.

I'm sorry I didn't answer your question directly: better with or without a counterweight.
This is the best answer I can give you.